Dual driver in non-small cell lung carcinoma – therapeutic dilemma

INTRODUCTION

The 5-year overall survival (OS) rate in stage IV non-small cell lung carcinoma (NSCLC) is <10%.^[1] Median progression-free survival (PFS) was 6–8 months and median OS was 10– 12 months with chemotherapy alone irrespective of epidermal growth factor receptor (EGFR)/ anaplastic lymphoma kinase (ALK)/C-Ros Oncogene 1 (ROS1) aberrations.^[2]

Activating EGFR mutations in lung adenocarcinoma constitute around 10–15% among Caucasians and around 50% among Asians.^[3] Most mutations in the EGFR gene are located within exons 19–21 of the tyrosine kinase domain. The rearrangement of ALK with echinoderm microtubule-associated protein-like 4 (EML4) oncogene on the chromosome 2p activates a specific tyrosine kinase, involved in the processes of survival and cell proliferation and is found in about 3–7% of lung adenocarcinomas. The ROS1 gene, that is present in the chromosome 6p encodes for a receptor tyrosine kinase can undergo rearrangement with multiple partner genes like CD74, EZG, FIG1, etc. renders the kinase constitutionally active. The ROS1 rearrangement can be seen in about 2% of the NSCLC population.^[4]

EGFR gene testing is done using hot spot mutation testing by polymerase chain reaction (PCR) and Sanger sequencing or by next-generation sequencing (NGS). The ALK and ROS1 gene screen testing is done by immunohistochemistry (IHC). The rearrangement of ALK and ROS1 genes are confirmed by fluorescence in-situ hybridization (FISH). The other genes which are commonly being tested for mutations are the BRAF, MET, and RET. These mutations can be targeted using various available drugs for specific genetic aberrations.

Generally among these driver mutations, only one of the mutations occurs in the tumor cells. On most occasions, these mutations are mutually exclusive. On very rare occasions, more than one driver mutation may coexist in the tumor. The most common reason for this co-existence is tumor heterogeneity. We describe five such cases of coexistence of two driver mutations (EGFR and ALK/ROS1 mutations).

CASE REPORT

DISCUSSION

Two different hypotheses exist to explain the presence of dual-driver mutations. The first hypothesis is that the genetic instabilities can cause genetic and phenotypic heterogeneity in the tumor, leading to different genetic alterations in different tumor cells rather than in a single clone of cells. The second hypothesis is that there can be activation of multiple oncogenic pathways due to alteration in a single clone of tumor cells.^[5] Such cases require molecular tumor board discussion for optimal management of patients.

The incidence of dual EGFR and ALK aberrations is about 1–5% in literature but conventional testing like the Sanger sequencing method has lower sensitivity to detect mutant cells.^[6-8] The targeted NGS has a sensitivity to detect <1% of mutants.^[9] More sensitive detection methods like NGS of the whole EGFR gene and mutant-enriched NGS have increased the rate of co-existence of EGFR mutation and ALK– translocations up to 15%.^[6]

Although EGFR mutations and ALK translocations are generally considered mutually exclusive, their concomitant prevalence differed due to direct sequencing for EGFR mutations. Their responses to EGFR and/or inhibitors were conflicting. Despite a response rate of 60–70% in EGFR-mutated lung cancer patients treated with EGFRTKIs, 10–20% of individuals developed primary resistance. Possible contributing factors are KRAS mutation, MET amplification, and PIK3CA mutation. The co-existence of a low burden EGFR mutation in the dual-positive patients might lead to the unfavorable response to EGFR-TKIs.^[10] The median PFS with the use of EGFR-TKIs in patients with EGFR mutations alone or with concomitant ALK/ROS1 aberrations were 10.7 and 6.6 months respectively.^[7] But there was no OS difference (23 months in both groups). Minor clone of EGFR mutant may have little influence on the responsiveness to ALK inhibitors in dual-positive patients. The presence of concomitant EGFR mutations has been regarded as a resistance mechanism of ALK inhibitors suggesting that there may be a need to use a combination therapy.^[6]

In a literature review of 100cases by Lo Russo et al., the disease control rates using EGFR (n = 53) and ALK (n = 39) TKIs were 69.8% versus 79.5% respectively and overall response rates (ORR) were 43.4 % and 51.3%, respectively. About 22 patients in this review series received EGFR TKIs followed by ALK TKIs. The ORR in this subgroup of 22 patients using EGFR TKI was 23.1% and ORR using ALK TKIs were 42.3%. Statistical comparisons and analysis of PFS and OS were not done in the above series.^[11] The responses to ALK TKIs were better than EGFR TKIs and this observation was similar to other series also.^[6,7]

Here in our series, the first three patients and case 5 were EGFR with ALK mutation and Case 4 was EGFR with ROS mutant. Regarding cases 1-3, the progression on EFGR TKIs happened in less than 6 months. The responses of the first 4 patients were either progressive disease or stable disease. We could also see that the responses were poor with EGFR TKIs which is similar to other series.^[6,7,11] In our series, we could not start any of our patients on ALK TKIs due to logistic issues.

Variant allele frequency (VAF) is the percentage of a specific sequence reads (DNA variant) observed divided by the overall coverage at that locus. VAF acts as a surrogate to measure the proportion of variant DNA molecules carried in the tumor biopsy/specimen.^[12] In a study by Friedlaender et al., high allelic frequency was significantly associated with PFS but not OS.^[13] When combined analysis of VAF and co-occurring mutations were performed those without co-mutations and with high VAF for EGFR mutation were classified as tumor sensitive to EGFR TKIs. They did significantly better in both PFS and OS. In a retrospective study by Gieszer et al., they calculated adjusted VAF (aVAF) for each patient. In this VAF was normalized to the proportion of neoplastic cells in each specimen. High EGFR-aVAF showed significant PFS and OS benefits compared to low EGFR-aVAF.^[14] These studies imply that therapy can be individualized based on VAF, especially when dual-driver mutations/heterogeneity exist depending on the mutation with high VAF. These hypotheses need to be validated in randomized control trials.

Over time, we shall come across more and more cases due to multiple reasons like the use of more sensitive methods for detection of mutations, the use of higher sensitive methods such as whole-genome sequencing,^[6] the use of expanded genetic panel testing. Identification of dual drivers per se may be identified as a factor for primary resistance soon. The optimal strategies for treatment in such cases either to sequence or to combine TKIs is not known yet. It requires more insight into molecular pathways or cross-talks between two driver pathways and more experience in treating such cases to identify appropriate treatment protocol.

CONCLUSION

Usage of highly sensitive wider molecular panels for testing in NSCLC will lead to a surge in the number of patients with dual drivers. These patients will require molecular tumour board discussion and a specialized approach for management. Further research is required in performing VAF analysis as a guide in the selection of drugs in such patients with dual drivers.